论文部分内容阅读
在色度、饱和度、纯度(HSV)彩色空间,结合简化脉冲耦合神经网络(S-PCNN)与二维离散静态小波(SWT)提出一种有效的遥感图像融合算法。将多色光谱转换到HSV色彩空间,对多色光谱的V分量与全色光谱进行二维静态小波分解,再将分解后的高频系数输入S-PCNN模型进行融合。低频部分进行第二次小波分解并采用不同规则将其融合,对融合的小波系数进行小波逆变换得到融合的V分量,并将多色光谱的H、S与融合后的V分量转换到RGB空间。通过一组常用的遥感图像融合实验,表明本文算法的融合效果优于传统算法,且融合图像细节明显、色彩保留较好,是一种有效的遥感图像融合算法。
An effective remote sensing image fusion algorithm based on simplified pulse coupled neural network (S-PCNN) and two-dimensional discrete static wavelet (SWT) is proposed in color, saturation and purity (HSV) The multi-color spectrum is converted to the HSV color space, the two-dimensional static wavelet decomposition of the V component and the full-color spectrum of the multi-color spectrum is performed, and the decomposed high frequency coefficient is input to the S-PCNN model for fusion. The low-frequency part of the second wavelet decomposition and the use of different rules to be fused, wavelet transform the fusion wavelet inverse transform to obtain the fused V component, and multi-color spectral H, S and fusion V component converted to RGB space . Through a set of commonly used remote sensing image fusion experiments, the proposed fusion algorithm is superior to the traditional one, and the fusion image has obvious details and good color retention. It is an effective remote sensing image fusion algorithm.